Skill Creator

zechenzhangAGI

75 published skills · 18,896 GitHub stars across the catalog. View on GitHub ↗

audiocraft-audio-generation

zechenzhangAGI's avatarfrom zechenzhangAGI★ 481

PyTorch library for audio generation including text-to-music (MusicGen) and text-to-sound (AudioGen). Use when you need to generate music from text descriptions, create sound effects, or perform melody-conditioned music generation.

blip-2-vision-language

zechenzhangAGI's avatarfrom zechenzhangAGI★ 481

Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text retrieval, or multimodal chat with state-of-the-art zero-shot performance.

segment-anything-model

zechenzhangAGI's avatarfrom zechenzhangAGI★ 481

Foundation model for image segmentation with zero-shot transfer. Use when you need to segment any object in images using points, boxes, or masks as prompts, or automatically generate all object masks in an image.

stable-diffusion-image-generation

zechenzhangAGI's avatarfrom zechenzhangAGI★ 481

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.

knowledge-distillation

zechenzhangAGI's avatarfrom zechenzhangAGI★ 481

Compress large language models using knowledge distillation from teacher to student models. Use when deploying smaller models with retained performance, transferring GPT-4 capabilities to open-source models, or reducing inference costs. Covers temperature scaling, soft targets, reverse KLD, logit distillation, and MiniLLM training strategies.

long-context

zechenzhangAGI's avatarfrom zechenzhangAGI★ 481

Extend context windows of transformer models using RoPE, YaRN, ALiBi, and position interpolation techniques. Use when processing long documents (32k-128k+ tokens), extending pre-trained models beyond original context limits, or implementing efficient positional encodings. Covers rotary embeddings, attention biases, interpolation methods, and extrapolation strategies for LLMs.

model-merging

zechenzhangAGI's avatarfrom zechenzhangAGI★ 481

Merge multiple fine-tuned models using mergekit to combine capabilities without retraining. Use when creating specialized models by blending domain-specific expertise (math + coding + chat), improving performance beyond single models, or experimenting rapidly with model variants. Covers SLERP, TIES-Merging, DARE, Task Arithmetic, linear merging, and production deployment strategies.

model-pruning

zechenzhangAGI's avatarfrom zechenzhangAGI★ 481

Reduce LLM size and accelerate inference using pruning techniques like Wanda and SparseGPT. Use when compressing models without retraining, achieving 50% sparsity with minimal accuracy loss, or enabling faster inference on hardware accelerators. Covers unstructured pruning, structured pruning, N:M sparsity, magnitude pruning, and one-shot methods.

moe-training

zechenzhangAGI's avatarfrom zechenzhangAGI★ 481

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

speculative-decoding

zechenzhangAGI's avatarfrom zechenzhangAGI★ 481

Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with limited compute. Covers draft models, tree-based attention, Jacobi iteration, parallel token generation, and production deployment strategies.

implementing-llms-litgpt

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen, Mistral). Use when need clean model implementations, educational understanding of architectures, or production fine-tuning with LoRA/QLoRA. Single-file implementations, no abstraction layers.

mamba-architecture

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

State-space model with O(n) complexity vs Transformers' O(n²). 5× faster inference, million-token sequences, no KV cache. Selective SSM with hardware-aware design. Mamba-1 (d_state=16) and Mamba-2 (d_state=128, multi-head). Models: 130M-2.8B on HuggingFace.

nanogpt

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

Educational GPT implementation in ~300 lines. Reproduces GPT-2 (124M) on OpenWebText. Clean, hackable code for learning transformers. By Andrej Karpathy. Perfect for understanding GPT architecture from scratch. Train on Shakespeare (CPU) or OpenWebText (multi-GPU).

rwkv-architecture

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux Foundation AI project. Production: Windows, Office, NeMo. RWKV-7 (March 2025). Models up to 14B parameters.

huggingface-tokenizers

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

Fast tokenizers optimized for research and production. Rust-based implementation tokenizes 1GB in <20 seconds. Supports BPE, WordPiece, and Unigram algorithms. Train custom vocabularies, track alignments, handle padding/truncation. Integrates seamlessly with transformers. Use when you need high-performance tokenization or custom tokenizer training.

sentencepiece

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

Language-independent tokenizer treating text as raw Unicode. Supports BPE and Unigram algorithms. Fast (50k sentences/sec), lightweight (6MB memory), deterministic vocabulary. Used by T5, ALBERT, XLNet, mBART. Train on raw text without pre-tokenization. Use when you need multilingual support, CJK languages, or reproducible tokenization.

axolotl

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

Expert guidance for fine-tuning LLMs with Axolotl - YAML configs, 100+ models, LoRA/QLoRA, DPO/KTO/ORPO/GRPO, multimodal support

llama-factory

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal support

unsloth

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization

nemo-curator

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features: fuzzy deduplication (16× faster), quality filtering (30+ heuristics), semantic deduplication, PII redaction, NSFW detection. Scales across GPUs with RAPIDS. Use for preparing high-quality training datasets, cleaning web data, or deduplicating large corpora.

ray-data

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

Scalable data processing for ML workloads. Streaming execution across CPU/GPU, supports Parquet/CSV/JSON/images. Integrates with Ray Train, PyTorch, TensorFlow. Scales from single machine to 100s of nodes. Use for batch inference, data preprocessing, multi-modal data loading, or distributed ETL pipelines.

grpo-rl-training

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

Expert guidance for GRPO/RL fine-tuning with TRL for reasoning and task-specific model training

openrlhf-training

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.

simpo-training

zechenzhangAGI's avatarfrom zechenzhangAGI★ 62

Simple Preference Optimization for LLM alignment. Reference-free alternative to DPO with better performance (+6.4 points on AlpacaEval 2.0). No reference model needed, more efficient than DPO. Use for preference alignment when want simpler, faster training than DPO/PPO.